Shulou(Shulou.com)11/24 Report--

CTOnews.com May 20 news, Nvidia has now released the RTX 4060 Ti 8G graphics card, which has only 128bit memory width. By contrast, Nvidia's older RTX 3060 Ti has a 256bit bit width and the RTX 3060 has a 192bit bit width.

Nvidia answered in a blog post that the storage subsystem of the new NVIDIA Ada Lovelace architecture increased the size of the L2 cache by 16 times compared to the previous generation GPU with 128bit display bit width, greatly improving the cache hit ratio.

As shown in the figure above, the L2 cache bandwidth in Ada GPU has increased significantly. This makes it possible to transfer more data between the processing core and the L2 cache.

In addition, Nvidia engineers tested a special test version of RTX 4060 Ti with 32 MB L2 cache and RTX 4060 Ti with only 2 MB L2.

In various games and comprehensive benchmarks, 32 MB level 2 cache reduces video memory bus traffic by an average of more than 50% compared to the performance of 2 MB level 2 cache. This 50% reduction in traffic enables GPU to use its video memory bandwidth more efficiently, with a nearly two-fold increase in efficiency. Therefore, in this case, the performance of the isolated video memory performance of Ada GPU with 288 GB / s peak video memory bandwidth is similar to that of Ampere GPU with 554 GB / s peak video memory bandwidth. In a series of games and comprehensive tests, the greatly improved cache hit rate increased the game frame rate by up to 34%.

Nvidia said that historically, the display bit width has been used as an important indicator to determine the speed and performance level of the new GPU. However, the explicit memory bit width itself does not fully indicate the performance of the storage subsystem. On the contrary, it is helpful to have a more comprehensive understanding of the storage subsystem design and its overall impact on game performance.

Due to the advantages of the Ada architecture, including the new RT and Tensor cores, higher clock speeds, the new OFA engine, and Ada's DLSS 3 features, the GeForce RTX 4060 Ti is faster and consumes less power than the previous generation of GeForce RTX 3060 Ti and RTX 2060 SUPER graphics cards.

CTOnews.com collates the parameter information of RTX 4060 Ti and RTX 4060 graphics cards as follows:

The RTX 4060 Ti has a 4352 CUDA core, equipped with 8GB / 16GB 128bit GDDR6 video memory, 160W / 165W TGP power consumption and a PCIe 4.0 connection. The price starts at 3199 yuan and goes on sale on May 24th.

The RTX 4060 graphics card has a 3072 CUDA core, is equipped with 8GB GDDR6 128bit memory, consumes 115W, and is connected with PCIe 4.0x8. The price starts at 2399 yuan and goes on sale in July.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.